CN104780310A - Image blurring detection method and system and camera - Google Patents
Image blurring detection method and system and camera Download PDFInfo
- Publication number
- CN104780310A CN104780310A CN201510040814.2A CN201510040814A CN104780310A CN 104780310 A CN104780310 A CN 104780310A CN 201510040814 A CN201510040814 A CN 201510040814A CN 104780310 A CN104780310 A CN 104780310A
- Authority
- CN
- China
- Prior art keywords
- image
- qcif
- fuzzy
- sense cycle
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Studio Devices (AREA)
Abstract
The invention is suitable for the field of image processing, and provides an image blurring detection method and system and a camera. The method comprises the following steps: extracting a frame of image from a video at a certain preset time interval in a preset detection period; performing down sampling on the images once a frame of image is extracted to obtain QCIF (Quarter Common Interchange Format) images; analyzing all the QCIF images in a preset detection period to obtain a value for judging whether or not the images are blurring; and determining whether or not the images of the video are blurring according to the value for judging whether or not the images are blurring. The operation amount is small, and great convenience is brought to algorithm implementation.
Description
Technical field
The invention belongs to image processing field, particularly relate to a kind of image blurring detection method, system and video camera.
Background technology
Now high-end video camera all has image blurring measuring ability.Safe city supervisory control system due to video camera too much; therefore often can run into a problem, a certain video camera is subject to abnormal change due to focal length, causes image blurring; often need several days even a few week time be just found, then image is during this period of time all fuzzy invalid.
Image blurring detection algorithm can be reported to the police in the time very short after fuzzy generation.Such as: in several seconds or a few minutes.Substantially increase the management maintenance of safe city supervisory control system to thousands of video camera.
The common image blurring detection method of prior art mainly comprises following two kinds:
1) by the method such as Fourier transform or wavelet transformation, in frequency domain, image is analyzed.
2) on the basis of rim detection, peak value and the wide association of the gradient of edge calculation judge.
But above-mentioned two kinds of method operands are all comparatively large, algorithm realization is very complicated.
Summary of the invention
The object of the present invention is to provide a kind of image blurring detection method, system and video camera, the image blurring detection method operand being intended to solve prior art is comparatively large, the problem that algorithm realization is very complicated.
First aspect, the invention provides a kind of image blurring detection method, described method comprises:
From video, a two field picture is extracted every a scheduled time in default sense cycle;
Often extract a two field picture then to carry out down-sampling to described image and obtain QCIF image;
All QCIF images in default sense cycle being analyzed, obtaining for judging the value whether image is fuzzy;
According to described for judging that the value whether image is fuzzy determines that whether the image of described video is fuzzy.
Second aspect, the invention provides a kind of image blurring detection system, described system comprises:
Abstraction module, for extracting a two field picture every a scheduled time in default sense cycle from video;
Down sample module, then carries out down-sampling to described image obtain QCIF image for often extracting a two field picture;
Analysis module, for analyzing all QCIF images in default sense cycle, obtains for judging the value whether image is fuzzy;
Determination module, for according to described for judging that the value whether image fuzzy determines that whether the image of described video fuzzy.
The third aspect, the invention provides a kind of video camera comprising above-mentioned image blurring detection system.
In the present invention, owing to extracting a two field picture every a scheduled time from video in default sense cycle; Often extract a two field picture then to carry out down-sampling to described image and obtain QCIF image; All QCIF images in default sense cycle being analyzed, obtaining for judging the value whether image is fuzzy; According to described for judging that the value whether image is fuzzy determines that whether the image of described video is fuzzy, therefore operand is lower, and algorithm realization is very convenient.
Accompanying drawing explanation
Fig. 1 is the flow chart of the image blurring detection method that the embodiment of the present invention one provides.
Fig. 2 is the structural representation of the image blurring detection system that the embodiment of the present invention two provides.
Embodiment
In order to make object of the present invention, technical scheme and beneficial effect clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.
In order to technical solutions according to the invention are described, be described below by specific embodiment.
embodiment one:
Refer to Fig. 1, the image blurring detection method that the embodiment of the present invention one provides comprises the following steps:
S101, in default sense cycle, from video, extract a two field picture every a scheduled time;
Within such as 30 seconds, in order to preset sense cycle, the scheduled time is 1 second, then S101 is specially: with 30 seconds for the one-time detection cycle, per secondly in 30 seconds from video, extracts a two field picture.
S102, often extract a two field picture and then down-sampling is carried out to described image and obtain QCIF image;
In the embodiment of the present invention one, S102 specifically comprises the following steps:
Often extract a two field picture and then down-sampling is carried out to described image, obtain CIF (Common Intermediate Format, common intermediate format) image; The image extracted can be high-definition image, also can be SD image; In H.323 protocol family, define CIF=352 × 288 pixel;
Noise reduction process is carried out and down-sampling to CIF image, obtains QCIF (Quarter CommonIntermediate Format, 1/4th common intermediate format) image; In H.323 protocol family, define QCIF=176 × 144 pixel.
S103, all QCIF images in default sense cycle being analyzed, obtaining for judging the value whether image is fuzzy;
S104, according to described for judging that the value whether image fuzzy determines that whether the image of described video fuzzy.
In the embodiment of the present invention one, S103 specifically can comprise the following steps:
The quantity N of the pixel suddenlyd change in statistics each QCIF image in default sense cycle
1;
In all QCIF images of statistics in default sense cycle, the quantity N of the pixel of sudden change
1be less than or equal to the number of times T of predetermined value n
n.Wherein n is natural number; T
nrepresent in default sense cycle, meet the quantity N of the pixel of sudden change
1be less than or equal to the number of times of predetermined value n, use different threshold value n to have different statisticses, such as T
50refer to N
1the number of times of <50, T
20refer to N
1the number of times of <20, T
0refer to N
1the number of times of=0.
Accordingly, S104 is specifically as follows:
Work as T
50>10 or T
20>5 or T
0during >2, determine that the image of described video is fuzzy.
In the embodiment of the present invention one, S103 specifically also can comprise the following steps:
The quantity N of the pixel suddenlyd change in statistics each QCIF image in default sense cycle
1;
In statistics each QCIF image in default sense cycle, differential intensity reaches the quantity N of the pixel of predetermined threshold
2;
By formula nRatio=N
2/ (N
1+ 1) multiple nRatio, is calculated; Under general case, the numerical value of nRatio is 5,6 or 7, then clear picture, and the numerical value of nRatio is more than or equal to 8, then image blurring;
In all QCIF images of statistics in default sense cycle, the number of times T of nRatio > r
r, wherein r is natural number; Different threshold value r is used to have different statisticses, such as T
8it is the number of times of nRatio > 8.
Accordingly, S104 is specifically as follows: work as T
8during >10, determine that the image of described video is fuzzy.
In the embodiment of the present invention one, S103 specifically also can comprise the following steps:
Layered image isolated pixel point quantity in statistics each QCIF image in default sense cycle;
In all QCIF images of statistics in default sense cycle, layered image isolated pixel point quantity is not the number of times of 0.
Accordingly, S104 is specifically as follows:
When the number of times that layered image isolated pixel point quantity is not 0 equals 0, determine that the image of described video is fuzzy.
The S103 of the embodiment of the present invention one can be above-mentioned three one of them schemes comprising the scheme of concrete steps, or above-mentioned three combination in any comprising the scheme of concrete steps, as long as the condition met in any one scheme of the concrete scheme included by above-mentioned S104 can determine that the image of described video is fuzzy.
In the embodiment of the present invention one, add up the quantity N of the pixel suddenlyd change in each the QCIF image in default sense cycle
1specifically comprise the following steps:
If following condition is set up, then this pixel is the pixel of sudden change, N
1counting adds 1.
As nV_L>T and nV_R>T time; Or
As nV_L<-T and nV_R<-T time; Or
As nV_T>T and nV_B>T time; Or
As nV_T<-T and nV_B<-T time; Or
As nV_TL>T and nV_BR>T time; Or
As nV_TL<-T and nV_BR<-T time; Or
As nV_TR>T and nV_BL>T time; Or
As nV_TR<-T and nV_BL<-T time
Wherein T represents threshold value, and T is natural number,
NV_L=p0-p8; Represent the difference of central pixel point and leftmost pixel point.
NV_R=p0-p4; Represent the difference of central pixel point and right pixels point.
NV_T=p0-p2; Represent the difference of central pixel point and top pixel.
NV_B=p0-p6; Represent the difference of central pixel point and following pixel.
NV_TL=p0-p1; Represent the difference of central pixel point and top left corner pixel point.
NV_BR=p0-p5; Represent the difference of central pixel point and lower right corner pixel.
NV_TR=p0-p3; Represent the difference of central pixel point and upper right corner pixel.
NV_BL=p0-p7; Represent the difference of central pixel point and lower left corner pixel.
P0, p1, p2, p3, p4, p5, p6, p7, p8 represent value, the value of top left corner pixel point, the value of top pixel, the value of upper right corner pixel, the value of right pixels point, value, the value of lower left corner pixel, the value of leftmost pixel point of the value of lower right corner pixel, below pixel of central pixel point respectively.
In the embodiment of the present invention one, add up differential intensity in each the QCIF image in default sense cycle and reach the quantity N of the pixel of predetermined threshold
2specifically comprise the following steps:
If following condition is set up, then the differential intensity of this pixel reaches predetermined threshold, N
2counting adds 1.
As nV_L>T; Or
As nV_L<-T; Or
As nV_R>T; Or
As nV_R<-T; Or
As nV_T>T; Or
As nV_T<-T; Or
As nV_B>T; Or
As nV_B<-T
In the embodiment of the present invention one, add up the measurer body of counting of layered image isolated pixel in each the QCIF image in default sense cycle and comprise the following steps:
Each QCIF image is pressed gray value, is divided into multiple rank (n rank);
By the pixel of each QCIF image according to tonal gradation again assignment, obtain the image of gray scale layering; Gray scale from 0 ~ 255, to be divided into 8 grades: 0 ~ 31 for grade 0, and gray scale is set to 0; 32 ~ 63 is grade 1, and gray scale is set to 32; 64 ~ 95 is grade 2, and gray scale is set to 64; 224 ~ 255 is grade 7, and gray scale is set to 224;
In the image of layering, find isolated pixel point, and count.
Wherein, the determination methods of isolated pixel point is: if the difference of all 8 pixels of pixel and 8 neighborhoods is all greater than threshold value T, be then considered to an isolated pixel point.Wherein T is natural number;
In the embodiment of the present invention one, owing to extracting a two field picture every a scheduled time from video in default sense cycle; Often extract a two field picture then to carry out down-sampling to described image and obtain QCIF image; All QCIF images in default sense cycle being analyzed, obtaining for judging the value whether image is fuzzy; According to described for judging that the value whether image is fuzzy determines that whether the image of described video is fuzzy, therefore operand is lower, and algorithm realization is very convenient.
embodiment two:
Refer to Fig. 2, the image blurring detection system that the embodiment of the present invention two provides comprises: abstraction module 11, down sample module 12, analysis module 13 and determination module 14, wherein,
Abstraction module 11, for extracting a two field picture every a scheduled time in default sense cycle from video;
Down sample module 12, then carries out down-sampling to described image obtain QCIF image for often extracting a two field picture;
Analysis module 13, for analyzing all QCIF images in default sense cycle, obtains for judging the value whether image is fuzzy;
Determination module 14, for according to described for judging that the value whether image fuzzy determines that whether the image of described video fuzzy.
In the embodiment of the present invention two, described analysis module 12 specifically can comprise:
First down sample module, then carrying out down-sampling to described image for often extracting a two field picture, obtaining CIF image;
Second down sample module, for carrying out noise reduction process and down-sampling to CIF image, obtains QCIF image.
In the embodiment of the present invention two, described analysis module 13 specifically can comprise:
First statistical module, for adding up the quantity N of the pixel suddenlyd change in each the QCIF image in default sense cycle
1;
Second statistical module, for adding up in all QCIF images in default sense cycle, the quantity N of the pixel of sudden change
1be less than or equal to the number of times T of predetermined value n
n; Wherein n is natural number;
Accordingly, described determination module 14 specifically may be used for: as the quantity N of the pixel of sudden change
1be less than the number of times T of 50
50the quantity N of the pixel of >10, sudden change
1be less than the number of times T of 20
20the quantity N of the pixel of >5 or sudden change
1equal the number of times T of 0
0during >2, determine that the image of described video is fuzzy.
In the embodiment of the present invention two, described analysis module 13 specifically also can comprise:
3rd statistical module, for adding up the quantity N of the pixel suddenlyd change in each the QCIF image in default sense cycle
1;
4th statistical module, reaches the quantity N of the pixel of predetermined threshold for adding up differential intensity in each the QCIF image in default sense cycle
2;
Computing module, for passing through formula nRatio=N
2/ (N
1+ 1) multiple nRatio, is calculated;
5th statistical module, for adding up in all QCIF images in default sense cycle, the number of times T of nRatio > r
r, wherein r is natural number;
Accordingly, described determination module 14 specifically may be used for: as the number of times T of nRatio > 8
8during >10, determine that the image of described video is fuzzy.
In the embodiment of the present invention two, described analysis module 13 specifically also can comprise:
6th statistical module, for adding up layered image isolated pixel point quantity in each the QCIF image in default sense cycle;
7th statistical module, for adding up in all QCIF images in default sense cycle, layered image isolated pixel point quantity is not the number of times of 0;
Accordingly, described determination module 14 specifically may be used for: when the number of times that layered image isolated pixel point quantity is not 0 equals 0, determine that the image of described video is fuzzy.
In the embodiment of the present invention two, the first statistical module and the 3rd statistical module specifically for: if following condition is set up, then this pixel is the pixel of sudden change, N
1counting adds 1;
As nV_L>T and nV_R>T time; Or
As nV_L<-T and nV_R<-T time; Or
As nV_T>T and nV_B>T time; Or
As nV_T<-T and nV_B<-T time; Or
As nV_TL>T and nV_BR>T time; Or
As nV_TL<-T and nV_BR<-T time; Or
As nV_TR>T and nV_BL>T time; Or
As nV_TR<-T and nV_BL<-T time
Wherein T represents threshold value, and T is natural number,
NV_L=p0-p8; Represent the difference of central pixel point and leftmost pixel point.
NV_R=p0-p4; Represent the difference of central pixel point and right pixels point.
NV_T=p0-p2; Represent the difference of central pixel point and top pixel.
NV_B=p0-p6; Represent the difference of central pixel point and following pixel.
NV_TL=p0-p1; Represent the difference of central pixel point and top left corner pixel point.
NV_BR=p0-p5; Represent the difference of central pixel point and lower right corner pixel.
NV_TR=p0-p3; Represent the difference of central pixel point and upper right corner pixel.
NV_BL=p0-p7; Represent the difference of central pixel point and lower left corner pixel.
P0, p1, p2, p3, p4, p5, p6, p7, p8 represent value, the value of top left corner pixel point, the value of top pixel, the value of upper right corner pixel, the value of right pixels point, value, the value of lower left corner pixel, the value of leftmost pixel point of the value of lower right corner pixel, below pixel of central pixel point respectively.
In the embodiment of the present invention two, the 4th statistical module specifically may be used for: if following condition is set up, then the differential intensity of this pixel reaches predetermined threshold, N
2counting adds 1;
As nV_L>T; Or
As nV_L<-T; Or
As nV_R>T; Or
As nV_R<-T; Or
As nV_T>T; Or
As nV_T<-T; Or
As nV_B>T; Or
As nV_B<-T
In the embodiment of the present invention two, the 6th statistical module specifically can comprise:
Divide module equally, for each QCIF image is pressed gray value, be divided into multiple rank (n rank);
Assignment module, for by the pixel of each QCIF image according to tonal gradation again assignment, obtain the image of gray scale layering; Gray scale from 0 ~ 255, to be divided into 8 grades: 0 ~ 31 for grade 0, and gray scale is set to 0; 32 ~ 63 is grade 1, and gray scale is set to 32; 64 ~ 95 is grade 2, and gray scale is set to 64; 224 ~ 255 is grade 7, and gray scale is set to 224;
Counting module, for finding isolated pixel point in the image of layering, and counts.
Wherein, the determination methods of isolated pixel point is: if the difference of all 8 pixels of pixel and 8 neighborhoods is all greater than threshold value T, be then considered to an isolated pixel point.Wherein T is natural number;
The embodiment of the present invention additionally provides a kind of video camera comprising the image blurring detection system that the embodiment of the present invention two provides.
One of ordinary skill in the art will appreciate that all or part of step realized in above-described embodiment method is that the hardware that can carry out instruction relevant by program has come, described program can be stored in a computer read/write memory medium, described storage medium, as ROM/RAM, disk, CD etc.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1. an image blurring detection method, is characterized in that, described method comprises:
From video, a two field picture is extracted every a scheduled time in default sense cycle;
Often extract a two field picture then to carry out down-sampling to described image and obtain QCIF image;
All QCIF images in default sense cycle being analyzed, obtaining for judging the value whether image is fuzzy;
According to described for judging that the value whether image is fuzzy determines that whether the image of described video is fuzzy.
2. the method for claim 1, is characterized in that, the described two field picture that often extracts then carries out down-sampling to described image and obtains QCIF image and specifically comprise:
Often extract a two field picture and then down-sampling is carried out to described image, obtain CIF image;
Noise reduction process is carried out and down-sampling to CIF image, obtains QCIF image.
3. method as claimed in claim 1 or 2, is characterized in that, describedly analyzes all QCIF images in default sense cycle, obtains for judging that the value whether image is fuzzy specifically comprises:
The quantity N of the pixel suddenlyd change in statistics each QCIF image in default sense cycle
1;
In all QCIF images of statistics in default sense cycle, the quantity N of the pixel of sudden change
1be less than or equal to the number of times T of predetermined value n
n; Wherein n is natural number;
For judging that value that whether image is fuzzy is determined that whether the image of described video is fuzzy and is specially described in described basis:
As the quantity N of the pixel of sudden change
1be less than the number of times T of 50
50the quantity N of the pixel of >10, sudden change
1be less than the number of times T of 20
20the quantity N of the pixel of >5 or sudden change
1equal the number of times T of 0
0during >2, determine that the image of described video is fuzzy.
4. method as claimed in claim 1 or 2, is characterized in that, describedly analyzes all QCIF images in default sense cycle, obtains for judging that the value whether image is fuzzy specifically comprises:
The quantity N of the pixel suddenlyd change in statistics each QCIF image in default sense cycle
1;
In statistics each QCIF image in default sense cycle, differential intensity reaches the quantity N of the pixel of predetermined threshold
2;
By formula nRatio=N
2/ (N
1+ 1) multiple nRatio, is calculated;
In all QCIF images of statistics in default sense cycle, the number of times T of nRatio > r
r, wherein r is natural number;
For judging that value that whether image is fuzzy is determined that whether the image of described video is fuzzy and is specially described in described basis:
As the number of times T of nRatio > 8
8during >10, determine that the image of described video is fuzzy.
5. method as claimed in claim 1 or 2, is characterized in that, describedly analyzes all QCIF images in default sense cycle, obtains for judging that the value whether image is fuzzy specifically comprises:
Layered image isolated pixel point quantity in statistics each QCIF image in default sense cycle;
In all QCIF images of statistics in default sense cycle, layered image isolated pixel point quantity is not the number of times of 0;
For judging that value that whether image is fuzzy is determined that whether the image of described video is fuzzy and is specially described in described basis:
When the number of times that layered image isolated pixel point quantity is not 0 equals 0, determine that the image of described video is fuzzy.
6. an image blurring detection system, is characterized in that, described system comprises:
Abstraction module, for extracting a two field picture every a scheduled time in default sense cycle from video;
Down sample module, then carries out down-sampling to described image obtain QCIF image for often extracting a two field picture;
Analysis module, for analyzing all QCIF images in default sense cycle, obtains for judging the value whether image is fuzzy;
Determination module, for according to described for judging that the value whether image fuzzy determines that whether the image of described video fuzzy.
7. system as claimed in claim 6, it is characterized in that, described analysis module specifically comprises:
First statistical module, for adding up the quantity N of the pixel suddenlyd change in each the QCIF image in default sense cycle
1;
Second statistical module, for adding up in all QCIF images in default sense cycle, the quantity N of the pixel of sudden change
1be less than or equal to the number of times T of predetermined value n
n; Wherein n is natural number;
Described determination module specifically for: as the quantity N of pixel of sudden change
1be less than the number of times T of 50
50the quantity N of the pixel of >10, sudden change
1be less than the number of times T of 20
20the quantity N of the pixel of >5 or sudden change
1equal the number of times T of 0
0during >2, determine that the image of described video is fuzzy.
8. system as claimed in claim 6, it is characterized in that, described analysis module specifically comprises:
3rd statistical module, for adding up the quantity N of the pixel suddenlyd change in each the QCIF image in default sense cycle
1;
4th statistical module, reaches the quantity N of the pixel of predetermined threshold for adding up differential intensity in each the QCIF image in default sense cycle
2;
Computing module, for passing through formula nRatio=N
2/ (N
1+ 1) multiple nRatio, is calculated;
5th statistical module, for adding up in all QCIF images in default sense cycle, the number of times T of nRatio > r
r, wherein r is natural number;
Described determination module specifically for: as the number of times T of nRatio > 8
8during >10, determine that the image of described video is fuzzy.
9. system as claimed in claim 6, it is characterized in that, described analysis module specifically comprises:
6th statistical module, for adding up layered image isolated pixel point quantity in each the QCIF image in default sense cycle;
7th statistical module, for adding up in all QCIF images in default sense cycle, layered image isolated pixel point quantity is not the number of times of 0;
Described determination module specifically for: when the number of times that layered image isolated pixel point quantity is not 0 equals 0, determine that the image of described video is fuzzy.
10. one kind comprises the video camera of the image blurring detection system described in any one of claim 6 to 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510040814.2A CN104780310A (en) | 2015-01-27 | 2015-01-27 | Image blurring detection method and system and camera |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510040814.2A CN104780310A (en) | 2015-01-27 | 2015-01-27 | Image blurring detection method and system and camera |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104780310A true CN104780310A (en) | 2015-07-15 |
Family
ID=53621526
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510040814.2A Pending CN104780310A (en) | 2015-01-27 | 2015-01-27 | Image blurring detection method and system and camera |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104780310A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111246203A (en) * | 2020-01-21 | 2020-06-05 | 上海悦易网络信息技术有限公司 | Camera blur detection method and device |
CN112040188A (en) * | 2020-09-01 | 2020-12-04 | 深圳市天一智联科技有限公司 | Method and device for identifying falling direction of user based on helmet and computer equipment |
CN114422713A (en) * | 2022-03-29 | 2022-04-29 | 湖南航天捷诚电子装备有限责任公司 | Image acquisition and intelligent interpretation processing device and method |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2003615A1 (en) * | 2007-06-12 | 2008-12-17 | Siemens Building Technologies Fire & Security Products GmbH & Co. oHG | Method and device for automatic assessment of the image quality of video images |
CN102378041A (en) * | 2011-10-25 | 2012-03-14 | 中兴通讯股份有限公司 | Image quality detection method and image quality detection device |
CN102598687A (en) * | 2009-10-10 | 2012-07-18 | 汤姆森特许公司 | Method and device for calculating blur in video images |
CN103093178A (en) * | 2011-10-28 | 2013-05-08 | 浙江大华技术股份有限公司 | Detection method of video image definition and moving range |
-
2015
- 2015-01-27 CN CN201510040814.2A patent/CN104780310A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2003615A1 (en) * | 2007-06-12 | 2008-12-17 | Siemens Building Technologies Fire & Security Products GmbH & Co. oHG | Method and device for automatic assessment of the image quality of video images |
CN102598687A (en) * | 2009-10-10 | 2012-07-18 | 汤姆森特许公司 | Method and device for calculating blur in video images |
CN102378041A (en) * | 2011-10-25 | 2012-03-14 | 中兴通讯股份有限公司 | Image quality detection method and image quality detection device |
CN103093178A (en) * | 2011-10-28 | 2013-05-08 | 浙江大华技术股份有限公司 | Detection method of video image definition and moving range |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111246203A (en) * | 2020-01-21 | 2020-06-05 | 上海悦易网络信息技术有限公司 | Camera blur detection method and device |
CN112040188A (en) * | 2020-09-01 | 2020-12-04 | 深圳市天一智联科技有限公司 | Method and device for identifying falling direction of user based on helmet and computer equipment |
CN112040188B (en) * | 2020-09-01 | 2022-07-08 | 深圳市天一智联科技有限公司 | Method and device for identifying falling direction of user based on helmet and computer equipment |
CN114422713A (en) * | 2022-03-29 | 2022-04-29 | 湖南航天捷诚电子装备有限责任公司 | Image acquisition and intelligent interpretation processing device and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108230357B (en) | Key point detection method and device, storage medium and electronic equipment | |
US10440390B2 (en) | Rapid selection method for video intra prediction mode and apparatus thereof | |
WO2014044158A1 (en) | Identification method and device for target object in image | |
CN110599453A (en) | Panel defect detection method and device based on image fusion and equipment terminal | |
CN110060278B (en) | Method and device for detecting moving target based on background subtraction | |
CN104680504A (en) | Scene change detection method and device thereof | |
CN113763270B (en) | Mosquito noise removing method and electronic equipment | |
JP5939023B2 (en) | Computer program and image extraction apparatus | |
DE102018109276A1 (en) | SCREENING BASIC SUPPORT FOR DYNAMIC LIGHTING SCENARIOS | |
CN111046862A (en) | Character segmentation method and device and computer readable storage medium | |
CN104780310A (en) | Image blurring detection method and system and camera | |
CN105979283A (en) | Video transcoding method and device | |
WO2023284609A1 (en) | Graphic code identification method and apparatus, computer device, and storage medium | |
CN110991437B (en) | Character recognition method and device, training method and device for character recognition model | |
CN111369557A (en) | Image processing method, image processing device, computing equipment and storage medium | |
CN114429479A (en) | Remote sensing image segmentation method and device based on image segmentation network | |
CN117576632A (en) | Multi-mode AI large model-based power grid monitoring fire early warning system and method | |
CN105913427B (en) | Machine learning-based noise image saliency detecting method | |
CN104602096A (en) | Detecting method and device for video subtitle area | |
CN110458203B (en) | Advertisement image material detection method | |
JPWO2021019830A5 (en) | ||
CN111311584A (en) | Video quality evaluation method and device, electronic equipment and readable medium | |
CN110363766A (en) | Mosaic detection method, device and electronic equipment | |
Banterle et al. | Real-Time High Fidelity Inverse Tone Mapping for Low Dynamic Range Content. | |
CN116486340A (en) | Foreign matter detection method and device for conveyor belt, storage medium and processor |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
EXSB | Decision made by sipo to initiate substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150715 |